Moving Into Using Data to Boost Compliance? Apply the ‘Snowball Method’

illustration of the snowball effect
Rene Perez Photo
Financial Crimes Consultant
Jack Henry

5 minutes

Starting small and going step-by-step can help credit union teams leverage analytics to deliver useful insights.

A significant opportunity for credit unions today is unlocking their treasure trove of data through in-depth analytics, and then organizing and normalizing it within their institutions. It can provide deeper context to member behaviors that extends beyond servicing current needs to predicting future behaviors with more personalization, protection and brand connectivity.

This is critical both in and out of a pandemic, but the current landscape demonstrates the necessity of this work. Overnight, the number of card-not-present versus card-present transactions skyrocketed as consumer behavior shifted from in-store to primarily online. U.S. online sales reached $63 billion in August, a 42% year-over-year increase, according to data from Adobe Analytics. The spike renders most data fraud models almost completely ineffective as they have been programmed to consumer’s specific habits, which have now changed. And per research from CUES Supplier member and strategic provider Cornerstone Advisors, Scottsdale, Arizona, over the last three years, credit unions have seen their collective share of new account applications drop by half from 51% to 25%.

Not only are credit unions tasked with growing the business while constantly facing the challenge to adequately comply with AML/BSA regulations, but now they must balance the long-term implications brought on by the pandemic. Due to silos and a lack of resources, credit union fraud and compliance teams often lack the data acumen to protect their institutions and members, track fraudulent activity and service member needs in a modern, holistic format.

And, when getting started, fraud and compliance teams often bite off more than they can chew. Because of this, they miss key opportunities to bring data analytics into the compliance space and develop meaningful conclusions from it. The industry can no longer afford this lack of connectivity. It can segregate the member experience and expose credit unions to unnecessary fraud risks.

Author and consumer finance expert Dave Ramsey espouses the snowball method for paying off debt. Think of how kids start with a small snowball and grow it by pushing it around the yard, picking up more snow as they go. Suddenly their snowball is a snow boulder! Ramsey suggests consumers who want to pay off debt start by paying off the lowest balance and then move on to pay off the next lowest balance—and the effect will grow like a snowball being pushed around picking up momentum and size—until all the debt is cleared.

This same, simple concept can be applied to credit union compliance teams that are increasingly expected to use progressive data analytic models to deliver insights. These small, deliberate steps will help compliance teams take manageable action now while keeping the end goal in mind:

  • Start small. Machine learning and artificial intelligence can be big, scary words to compliance teams tasked with mitigating risk for the credit union. This tends to happen when teams start out trying to boil the ocean and quickly get overwhelmed with the technical complexities of these models. In the first step, the only requirement is to use what’s already available. Just like Ramsey encourages chipping away at the lowest debt first, compliance teams should do the same by analyzing a small sample of member data for a myriad of things that can predict future behaviors, including historical information, spending patterns and more.
  • Find the low-hanging fruit and eat it. Ramsey finds that consumers who experience small victories—like paying off a credit card—on their journeys to becoming debt-free tend to stay motivated and keep going. The same is true for data and compliance teams: Take the small wins. Once insights are gained from the small sample of data, apply the results to compliance and fraud models or create new models for more member transparency and behavioral knowledge. This information can then be shared across other disparate data teams for more collaboration and synchronization and for model development and strategies for programs designed to retain high-value members, market the right products, and gauge the risks of offering loans. This step helps compliance teams go from protecting and governing data to identifying and mitigating external and internal risks—and sharing those results for more value across the organization. It can be especially crucial during this time when traditional credit and behavior models offer less insight into member behaviors.  
  • Learn the lessons from each small project and, when taking on the bigger pieces, the plan will be stronger and field tested. Ramsey encourages consumers to roll over payments to the next larger debts. For compliance teams, once the first steps are in rotation and constantly being refined, it becomes easier to employ more advanced models for predicting member behavior. These progressive data analytic models then become standard for compliance teams for the reduction and/or elimination of cybercrimes, attacks on personal data, and the quick identification of subtle changes in member behaviors as they relate to quick decisions about financial transactions. Additionally, credit unions can then focus on which products they should invest in for the greatest return.

The techniques Ramsey lays out for consumers to tackle debt are easily transferable to data teams that want to create a smoother alternative for tackling a potentially complicated infrastructure, creating a process and normalizing it. These starting points grant access to progressive data analytics without a large barrier to entry. The results are also interesting to management and board members.

The lines along which departments manage data are blurring as quickly as digital experiences; now is the perfect to get a head start on more integration and knowledge-sharing between your compliance, analytics, fraud and cybersecurity groups. Together, progressive data analytics can have a profound impact on how credit unions are run, protected and drive revenue. It’s a necessary strategy for credit unions to move upstream and stay ahead of members in the ever-changing world we live in.

Rene Perez is a financial crimes consultant for Jack Henry (NASDAQ:JKHY), Monett, Missouri, a leading provider of technology solutions primarily for the financial services industry. He is a member of the Federal Reserve’s Fraud Definitions Work Group. Jack Henry is a S&P 500 company that serves approximately 8,700 clients nationwide.

CUES Learning Portal